Efficient and Accurate Discovery of Patterns in Sequence Data Sets

Abstract

Existing sequence mining algorithms mostly focus on mining for subsequences. However, a large class of applications, such as biological DNA and protein motif mining, require efficient mining of "approximate" patterns that are contiguous. The few existing algorithms that can be applied to find such contiguous approximate pattern mining have drawbacks like poor scalability, lack of guarantees in finding the pattern, and difficulty in adapting to other applications. In this paper, we present a new algorithm called FLAME (FLexible and Accurate Motif DEtector). FLAME is a flexible suffix tree based algorithm that can be used to find frequent patterns with a variety of definitions of motif (pattern) models. It is also accurate, as it always find the pattern if it exists. Using both real and synthetic datasets, we demonstrate that FLAME is fast, scalable, and outperforms existing algorithms on a variety of performance metrics. Using FLAME, it is now possible to mine datasets that would have been prohibitively difficult with existing tools.

ACME: A scalable parallel system for extracting frequent patterns from a very long sequence

“…Let D be a function that measures similarity between two sequences. Following the previous work [8,9], in this paper we assume D is the Hamming distance (i.e., number of mismatches A maximal motif must be right maximal and left maximal [8]. m is right maximal if L(mα) has less occurrences or more mismatches than L(m), where α ∈ Σ.…”

“…For example, node 1.2 in Fig. 3 is annotated with f = 2 because its path label TGC appears in S at (9,11) and (20,22). For clarity, we do not show the left-diversity annotation in the figure.…”

“…Intuitively, extracting maximal and supermaximal motifs is more complex than exact-length ones because, if length is known, the search space can be pruned significantly. FLAME [9] supports only exact-length motifs. To explore a sequence, users need to run multiple exactlength queries.…”

“…The statistical ones rely on sampling or calculating the probability of motif existence, and trade accuracy for speed [14]; they may miss motifs (false negatives) or return motifs that do not exist (false-positives). Combinatorial approaches [1,9,10], on the other hand, verify all combinations of symbols and return all motifs that satisfy the user's criteria. This paper focuses on the combinatorial case.…”

“…FLAME [9], for instance, searches for motifs of a specific length only. Despite this restriction, the largest reported input sequence is only 1.3 MB.…”